The Site Selection Problem

Before a single solar panel or wind turbine goes into the ground, developers face months of manual, fragmented analysis to find land that actually works. Engineers and analysts stitch together satellite imagery, land-use records, grid connection data, environmental constraints, and permitting rules from dozens of disconnected sources. The process is slow, expensive, and error-prone, and a single overlooked constraint can sink a project years into development. As renewable energy build-out accelerates worldwide, this bottleneck has become one of the biggest hidden costs in the industry.

What They're Building

Plume has built an AI-powered geospatial platform that automates the search for optimal renewable energy sites. The platform ingests satellite imagery, topography, grid infrastructure, land ownership, environmental data, and regulatory constraints, then uses machine learning to score and rank locations for solar and wind development in a fraction of the time it takes analysts to do it manually. Instead of weeks of manual GIS work, developers can screen entire regions and surface viable sites in minutes, complete with the underlying data needed to justify each recommendation.

By turning site selection into a data-driven, repeatable workflow, Plume lets renewable energy developers move faster from prospecting to permitting. The platform is designed to plug into existing development pipelines, giving in-house teams and independent power producers a shared, defensible source of truth on where projects are most likely to succeed.

Funding and Growth

Plume raised a €3.3 million (~$3.9 million) seed round to expand its geospatial AI platform and grow its engineering and go-to-market teams. The round was led by AENU, a climate-focused venture fund, with participation from Y Combinator and Kima Ventures. Plume is a graduate of Y Combinator's Summer 2024 batch, where it sharpened its product around the core insight that renewable energy developers were still picking sites the same way they did a decade ago — manually, and at great cost.